News for hackers. Stuff that matters.

A comment on The Hanabi Challenge: A New Frontier for AI Research. Parent comment

Re:The Hanabi Challenge: A New Frontier for AI Research by gamegoblin on Tuesday February 05, 2019 @02:35PM (#19088909)
When playing with casuals I actually don't mind communicating deductions that could strictly be made with a very simple application of the rules, such as your example. If someone discards a RED-4, there isn't a deep chain of logical reasoning to arrive at the conclusion that only 1 is left.

The best to teach casuals, I have found, is to play a round where you actually invert it have the experience player announce all of their reasoning (to demonstrate to the casuals the type of deductions that can be made).

[ Parent | on HN ]
Re:The Hanabi Challenge: A New Frontier for AI Research by defen on Tuesday February 05, 2019 @02:51PM (#19089073)
That's true, for teaching purposes it does help to simplify, as long as people know that they're not playing by the "real" rules (so when they play again without you, they play correctly). I don't remember the specific details, but the times I've seen it be an issue are when people give different answers to a question since they've each deduced different things about the board state.
[ Parent | on HN ]